Nisin, a peptide antibiotic, efficiently kills bacteria through a unique mechanism which includes inhibition of cell wall biosynthesis and pore formation in cytoplasmic membranes. Both mechanisms are based on interaction with the cell wall precursor lipid II which is simultaneously used as target and pore constituent. We combined two biosensor techniques to investigate the nisin activity with respect to membrane binding and pore formation in real time. Quartz crystal microbalance (QCM) allows the detection of nisin binding kinetics. The presence of 0.1 mol% lipid II strongly increased nisin binding affinity to DOPC (k(D) 2.68 x 10(-7) M vs. 1.03 x 10(-6) M) by a higher association rate. Differences were less pronounced while using negatively charged DOPG membranes. However, lipid II does not influence the absolute amount of bound nisin. Cyclic voltammetry (CV) data confirmed that in presence of 0.1 mol% lipid II, nanomolar nisin concentrations were sufficient to form pores, while micromolar concentrations were necessary in absence of lipid II. Both techniques suggested unspecific destruction of pure DOPG membranes by micromolar nisin concentrations which were prevented by lipid II. This model membrane stabilization by lipid II was confirmed by atomic force microscopy. Combined CV and QCM are valuable to interpret the role of lipid II in nisin activity.

Lantibiotics, a group of lanthionine-containing peptides, display their antibiotic activity by combining different killing mechanisms within one molecule. The prototype lantibiotic nisin was shown to possess both inhibition of peptidoglycan synthesis and pore formation in bacterial membranes by interacting with lipid II. Gallidermin, which shares the lipid II binding motif with nisin but has a shorter molecular length, differed from nisin in pore formation in several strains of bacteria. To simulate the mode of action, we applied cyclic voltammetry and quartz crystal microbalance to correlate pore formation with lipid II binding kinetics of gallidermin in model membranes. The inability of gallidermin to form pores in DOPC (1,2-dioleoyl-sn-glycero-3-phosphocholine) (C18/1) and DPoPC (1,2-dipalmitoleoyl-sn-glycero-3-phosphocholine) (C16/1) membranes was related to the membrane thickness. For a better simulation of bacterial membrane characteristics, two different phospholipids with branched fatty acids were incorporated into the DPoPC matrix. Phospholipids with methyl branches in the middle of the fatty acid chains favored a lipid II-independent DPoPC permeabilization by gallidermin, while long-branched phospholipids in which the branch is placed near the hydrophilic region induced an identical lipid II-dependent pore formation of gallidermin and nisin. Obviously, the branched lipids altered lipid packing and reduced the membrane thickness. Therefore, the duality of gallidermin activity (pore formation and inhibition of the cell wall synthesis) seems to be balanced by the bacterial membrane composition.

Selectins are adhesion receptors that participate in inflammation and tumor cell metastasis. The anti-inflammatory and antimetastatic activities of heparins have been related partly to their ability to interact with P- and L-selectin. The recent findings that various heparins differ in antimetastatic activity were explained by differences in their P- and L-selectin binding ability. To obtain data to illustrate the binding characteristics, we detected for the first time the binding kinetics and affinity of the two low molecular weight heparins (LMWHs) enoxaparin and nadroparin, and of the unfractionated heparin Liquemin N to P- and L-selectin using a quartz crystal microbalance biosensor. Enoxaparin and nadroparin behave nearly identical in their binding affinity to both P-selectin ( KD 4.60 x 10 (- 6) M versus 7.61 x 10 (- 6) M) and L-selectin ( KD 2.01 x 10 (- 6) M versus 2.84 x 10 (- 6) M). Liquemin N displayed slightly higher affinities to both selectins ( KD 6.07 x 10 (- 7) M versus 1.07 x 10 (- 7) M). The differences are caused by a higher association rate compared with that of the LMWHs. These data support recent findings of antimetastatic activities, but illustrate that the intrinsic selectin binding does not entirely reflect the antimetastatic activities in vivo.

Ketoprofen is an important anti-inflammatory drug, but its dermal application is limited because of the photosensitizing properties causing phototoxic reactions of the skin when exposed to UV light. We have recently investigated the peroxide formation of ketoprofen in solutions of linoleic acid during UV irradiation. To continue these studies and focus on UV-induced changes in membrane integrity and barrier function we established an in vitro model using two biosensor techniques simultaneously. Support-fixed bilayers were irradiated with different doses of UV-B up to damaging intensities with or without ketoprofen. Cyclic voltammetry was carried out to detect alterations in membrane permeability; quartz crystal microbalance (QCM) measurements were helpful in analyzing whether a permeability increase was caused by depletion of membrane components. In absence of ketoprofen, increasing UV-B doses induce membrane permeabilities of both unsaturated and saturated bilayers. QCM measurements could not reveal a significant loss of membrane components as a reason for the permeability. In contrast, 0.3 mM ketoprofen induced a dose-dependent increase in membrane permeability. QCM results indicated a mass loss. Although this model does not explain all molecular mechanisms of membrane damage by ketoprofen, the combined application of both QCM and CV is a novel and powerful tool to investigate functional mechanisms of UV-induced membrane damages.